Methods

1
The relationship between cortisol and memory
Preliminary analysis on the role of affective
state and the amygdala Allison L. Jahna,b, Simone
Kernb, Richard J. Davidsona, b, Jerry L.
Halversona, Clemens Kirschbaumc, and Heather C.
Abercrombiea University of Wisconsin-Madison
Departments of Psychology Psychiatrya, The
Waisman Laboratory for Brain Imaging and
Behaviorb, Technical University of Dresdenc

• Methods
• AVOTEC goggle system Eprime
  
  Presentation Software for stimuli
  display
• SRET (Self-Referent Encoding Task) participants
  
  asked to decide whether descriptive words
  are
  self-referent by
  making a dichotomous button press.
  
  Each scan has different sets of words matched
  
  for valence and arousal.
• Memory session recognition memory assessed 4-6
  days
  after scanning session

RESULTS Memory Data for Low High
PA Amygdala ROI for Low High PA Contrast
Negative Word Trials (Neutral Positive Word
Trials)
Background Historically, cortisol has been
considered harmful to cognition and health.
However, animal and human data suggest that
moderate cortisol elevations can be beneficial.
Cortisol release affects cognitive processes
such as memory. Research is divided with some
studies finding that cortisol impairs memory and
others finding that cortisol facilitates memory
(Maheu et al. 2004 Abercrombie et al. 2003).
Interestingly, research suggests that the
beneficial effects of cortisol on memory depend
on the emotional state of the individual (Okuda
et al., 2004 Abercrombie et al., 2006). These
studies suggest that state negative affect plays
a permissive role in the effects of cortisol on
memory. Animal data suggests that amygdala
activation, specifically the basolateral region
of the amygdala, is necessary for the beneficial
effects of cortisol on memory. To date, this is
the first human investigation of the role of the
amygdala in the effects of cortisol on memory.
Hypotheses We hypothesize that a moderate dose
of cortisol given in the late afternoon (when
endogenous levels are low) will be related to
increases in brain activation in the amygdala
compared to placebo. We hypothesize that this
effect will be associated with emotional state.
Finally, we hypothesize that changes in emotional
state will be related to changes in memory for
words encoded during elevated cortisol.
Beginning approx. 430pm
Significance p 6.16, p
Average Percent Signal Change Left Ventral
Amygdala
Time Course Left Ventral Amygdala

• Participants
• Data collection is ongoing
• 17 Participants (5 Women Age-M28.72 SD7.21
  3 participants dropped from analysis 1 for
  abnormal levels of cortisol on the cortisol scan
  day 1 experimenter error 1 abnormal memory
  values on the memory testing day)
• Right handed Free of general health issues
  Free of psychiatric disorders

Trend p
Conclusion While these analyses are preliminary,
we find support for increased brain activation in
the ventral amygdala after the administration of
cortisol only in individuals with low positive
affect. This effect was found in response to
negative words compared to neutral and positive.
Also, we find that individuals with low positive
affect also show greater memory for words encoded
during cortisol administration and that this
effect is more evident when examining memory for
negatively valenced words. Future directions We
hypothesize that cortisol will alter brain
activation in other regions such as the
hippocampus, parahippocampal gyrus, and
prefrontal cortex. Activation in these regions
may be especially important because of their
association with memory and emotion. We are also
collecting data on emotion and arousal ratings of
the stimuli in the scanner. Thus, we will examine
the relationship among memory, changes in brain
activation, and affective responses.
MRI Methods Analysis Quadrature Birdcage Coil
for T1 high resolution anatomical BOLD Sagital
Acquisition Flip Angle 90 degrees TR 2000 ms
TE 30 ms Slices per Volume 30 Voxel Size 3.75
x 3.75x 4mm with 1 mm gap AFNI Slice timing
correction, Motion correction, Spatial smoothing
6mm Normalization to Talairach space Extraction
of amygdala ROI in AFNI with predefined
masks GLM Gamma Variate Function percent signal
change from baseline for Negative Word Trials
(Neutral Positive Word Trials)

• Group Selection
• Positive and Negative Affect Schedule (Watson,
  Clark, Tellegen, 1988)
• Due to low variation in negative affect scores we
  used positive affect as an index of affective
  state
• Median split 7 Low Positive Affect (Low PA 3
  women) 7 High Positive Affect (High PA 2 women)

References Abercrombie HC, Kalin NH, Thurow ME,
Rosenkranz MA, Davidson RJ (2003). Cortisol
variation in humans affects memory for
emotionally laden and neutral information. Beh
Neuroscience 117505-16. Abercrombie, H.C.,
Speck, N.S., Monticelli, R.M. (2006).
Endogenouscortisol elevations are related to
memory facilitation only inindividuals who are
emotionally aroused. PNEC 31187-96. Maheu FS,
Joober R., Beaulieu S, Lupien SJ (2004).
Differential Effects of Adrenergic and
Corticosteroid Hormonal Systems on Human Short-
and Long-Term Declarative Memory for Emotionally
Arousing Material. Beh Neuroscience
118420-8. Okuda S, Roozendaal B, McGaugh JL
(2004). Glucocorticid effects on object
recognition memory require training-associated
emotional arousal. PNAS 101853-8. Watson D,
Clark LA, Tellegen A (1988). Development and
Validation of Brief Measures of Positive and
Negative Affect The PANAS Scales. J Personality
Social Psych 541063-70. This research was
supported in part by a NARSAD Young Investigator
Award given to Heather Abercrombie.